SYSTEMS AND METHODS FOR DETERMINING POWER CONSUMPTION PROFILES FOR RESOURCE USERS AND USING THE PROFILES FOR RESOURCE ALLOCATION

• US 20080184230A1
• Filed: 01/31/2007
• Published: 07/31/2008
• Est. Priority Date: 01/31/2007
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Leech, Phillip A., Baker, Dennis
• Granted Patent: US 8,037,329 B2
• Time in Patent Office: Days
• US Class Current: 718/100
• CPC Class Codes: G06F 1/329 by task scheduling
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5094 where the allocation takes into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

According to one embodiment, a system comprises a plurality of resources. The resources may be any limited-capacity computing resource that is capable of serving a computing demand of a resource user. A resource user refers to any entity which consumes computing capacity of a resource. The system comprises at least one power consumption profile associated with a respective resource user. The system also comprises at least one power management agent for measuring power consumption by ones of the plurality of resources in serving computing demands of the resource user, wherein said power consumption profile contains information based at least in part on the measured power consumption. The system also comprises at least one allocation agent for dynamically allocating the resource user to at least one of the resources based at least in part on the resource user'"'"'s power consumption profile.

Claims

1. A system comprising:

a plurality of resources;

at least one power consumption profile associated with a respective resource user;

at least one power management agent for measuring power consumption by ones of said plurality of resources in serving computing demands of said resource user, wherein said power consumption profile contains information based at least in part on the measured power consumption; and

at least one allocation agent for dynamically allocating said resource user to at least one of said resources based at least in part on said resource user'"'"'s power consumption profile.

2. The system of claim 1 wherein said plurality of resources comprise any limited-capacity computing resource that is capable of serving a computing demand of said resource user.

3. The system of claim 1 wherein said plurality of resources comprise at least one of the following:

blade computer, personal computer, printer, data storage resource, processing resource, network interface resource, and input/output resource.

4. The system of claim 1 wherein said resource user comprises any entity which consumes computing capacity of said at least one resource.

5. The system of claim 1 wherein said resource user comprises any of the following;

human user, hardware, and software process.

6. The system of claim 1 wherein said at least one power management agent updates said information in said power consumption profile based on historical measured power consumption by ones of said plurality of resources in serving computing demands of said resource user.

7. The system of claim 1 wherein said at least one allocation agent determines expected power consumption by said plurality of resources in serving computing demands of said resource user based at least in part on said resource user'"'"'s power consumption profile, and wherein said at least one allocation agent determines which one or more of said plurality of resources to allocate for serving said computing demands of said resource used based at least in part on the determined expected power consumption.

8. The system of claim 1 comprising a plurality of said resource users, further comprising;

for each of said plurality of resource users, at least one said power consumption profile associated with a respective one of said plurality of resource users; and

wherein said at least one allocation agent determines which one or more of said resources to allocate to each of said plurality of resource users based at least in part on the resource users'"'"' respective power consumption profile.

9. The system of claim 8 wherein said at least one allocation agent determines which one or more of said resources to allocate to each of said plurality of resource users concurrently present in the system.

10. The system of claim 1 wherein said power management agent determines a consumption concurrency value for said plurality of resources based at least in part on power consumption data collected from said plurality of resources.

11. The system of claim 10 wherein said power management agent determines whether said consumption concurrency value is greater than a concurrency threshold defined for said plurality of resources;

and wherein when said consumption concurrency value exceeds said defined concurrency threshold, said power management agent notifies said allocation agent.

12. A method comprising:

identifying a resource user having computing demands that can be serviced by any of a plurality of computing resources;

accessing a power consumption profile associated with said resource user;

determining, by an allocation agent, which one or more of said plurality of computing resources to allocate to said resource user based at least in part on said resource user'"'"'s power consumption profile; and

allocating, by said allocation agent, the determined one or more of said plurality of computing resources to said resource user.

13. The method of claim 12 further comprising:

monitoring, by a power management agent, power consumption of said resources in serving said computing demands of said resource user; and

updating information in said resource user'"'"'s power consumption profile to reflect a historical amount of power consumption by said resources attributable to serving said resource user'"'"'s computing demands.

14. The method of claim 13 further comprising:

determining, based at least in part on said historical amount of power consumption by said resources attributable to serving said resource user'"'"'s computing demands, all expected amount of power consumption by said resources for serving computing demands of said resource user.

15. The method of claim 14 wherein said determining which one or more of said plurality of computing resources to allocate to said resource user comprises:

determining which one or more of said plurality of computing resources to allocate to said resource user based at least in part on the determined expected amount of power consumption by said resources for serving computing demands of said resource user.

16. The method of claim 12 further comprising:

determining which one or more of said plurality of computing resources to allocate to said resource user based at least in part on other resource users'"'"' respective power consumption profiles.

17. A method of claim 12 further comprising:

determining a consumption concurrency value for a set of said plurality of resources based at least in part on power consumption data from said set of plurality of resources;

determining whether said consumption concurrency value is greater than said set of resources'"'"' concurrency threshold;

when said set of resources'"'"' said consumption concurrency value exceeds said concurrency threshold, sending a notification to said allocation agent.

18. A method comprising:

monitoring, by a power management agent, power consumption of computing resources in serving computing demands of resource users;

providing information in a respective power consumption profile associated with one of the resource users to reflect a historical amount of power consumption by said computing resources attributable to serving said computing demands of said one of the resource users;

determining, based at least in part on said historical amount of power consumption by said computing resources attributable to serving said computing demands of said one of the resource users, an expected amount of power consumption by said computing resources for serving computing demands of said one of the resource users;

determining which one or more of said computing resources to allocate to said one of the resource users based at least in part on the determined expected amount of power consumption by said computing resources; and

allocating, by an allocation agent, the determined one or more of said computing resources to said one of the resource users.

19. A method of claim 18 further comprising.

providing said power consumption profile for each of a plurality of said resource users, wherein the power consumption profile of each resource user includes information reflecting a historical amount of power consumption by said computing resources attributable to serving said computing demands of such resource user;

determining, for each of said resource users, based at least in part on their respective power consumption profiles, an expected amount of power consumption by said computing resources for serving respective computing demands of such resource user;

determining, for each of said resource users, which one or more of said computing resources to allocate to such resource user based at least in part on the determined expected amount of power consumption by said computing resources in serving the respective computing demands of such resource user.

20. The method of claim 19 further comprising:

determining, for each of said resource users, which one or more of said computing resources to allocate to such resource user based at least in part on the determined expected amount of power consumption by said computing resources for the resource users concurrently present in a system.